Very accurate. Early in the season it's tough to tell since not as many games played but after a a month, it gets pretty accurate at least at the 04 level.
Not very accurate at the beginning of the season. The algorithm works on averaging goal differential and strength of schedule. Right now, there is not enough games played or enough games to cross to determine strength of schedule. also, the average goal differential could be heavily influenced by a handful of games since the we are a couple weeks into the season. think of it as an expanding web of games and scores and each weekend, it will be more accurate. I would give it another month before it gets spot on accurate.
Not very accurate at the beginning of the season. The algorithm works on averaging goal differential and strength of schedule. Right now, there is not enough games played or enough games to cross to determine strength of schedule. also, the average goal differential could be heavily influenced by a handful of games since the we are a couple weeks into the season. think of it as an expanding web of games and scores and each weekend, it will be more accurate. I would give it another month before it gets spot on accurate.
Correct. Things will move a lot in the coming weeks, then settle in so that it is very difficult to move up or down once the database reflects a couple months of action. I would imagine that's why they introduced the "last 10" tab a couple years ago, because the rankings get boring late in the season because there is so little movement week to week.
In my experience, the ratings are pretty accurate as long as you realize they are not meant to crown a champion or to have the resolution to determine whether #10 is a better team than #11. They are really meant to show the relative strength of teams so that team managers can select tournaments that are at the appropriate level for their team, or to schedule "friendlies" with teams that will give them a good game.
That being said, if you play in a weak league, it is very difficult to maintain a high rating because your strength of schedule will suffer. This tends to deflate the rating of good E9 teams and can make them look weaker than mid-level EHF teams, who experience the opposite effect (inflated rating) because they play the top 2-3 teams in the area four times during the season. I haven't looked closely (probably because there are so few data points), but I would bet the top E9 teams outperform their rankings against the mid-low level EHF teams because of this effect.
BTW, this effect provides some incentive to EHF owners to withhold games against the highly rated EHF teams in the name of protecting the rankings of their franchises. It's very difficult for an E9 team to gain in the rankings without the opportunity to play against the top EHF teams and boost their strength of schedule metric.
Sorry for the long post, I love working the numbahs.
You need to get a life, or go buy a long rope and use it.
or you could get a life and let the guy have his say. who cares if he likes to run the numbers and talk about it. I am sure your Warcraft and Pokemon friends are waiting for you so move along.
You need to get a life, or go buy a long rope and use it.
Thank you my friend. We spend so much time arguing about teams and rankings that I think it's worth getting to understand how the MHR process works. What do you suggest I should have posted to a thread titled "My Hockey Rankings" which had already been seeded with a couple of thoughtful posts?
E9 top teams have their chance to improve in the rankings too - beat the weaker teams by 7 goals.
They might get dinged on the SoS rating, but you can make up for that with dominating wins.
Just have to avoid the 2-goal wins against bad teams.
Maybe that formula won't help you break the top-10, but you can hover inside the Top25 all year that way.
E9 top teams have their chance to improve in the rankings too - beat the weaker teams by 7 goals.
They might get dinged on the SoS rating, but you can make up for that with dominating wins.
Just have to avoid the 2-goal wins against bad teams.
Maybe that formula won't help you break the top-10, but you can hover inside the Top25 all year that way.
That is true, and a few do, but as you know these kids play up or down to their competition so there are very few 7 goal spreads. The goal is generally to win, not to win by 7.